U.S. Lost Jobs in September, ADP Says. Why the Picture May Not Be So Bleak.

Construction workers build homes at a new housing development in Henderson, Nev. (Justin Sullivan/Getty Images)

Key Points

• U.S. private payrolls shed 32,000 jobs last month, contrary to economists’ expectations of a 50,000 job gain.
• An annual preliminary benchmarking of ADP’s payroll data reduced September’s job count by 43,000 compared to pre-benchmarked data.
• The ADP report, based on over 26 million U.S. employees’ payroll data, offers insights into labor trends, especially during government shutdowns.

The latest data show that employment in the private sector slid last month, a troubling sign that will be harder to parse because the government shutdown means other job-market indicators aren’t likely to be available.

The U.S. shed 32,000 jobs from private payrolls last month, according to the ADP National Employment Report released Wednesday morning.

That is weaker than economists expected, although technical factors weighed on the numbers and other job-market indicators still appear less bleak. Those surveyed by FactSet estimated that the U.S. private employers would add 50,000 jobs in September. The data also showed a revised loss of 3,000 positions in August, down from an initially reported gain of 54,000 jobs.

The numbers, showing a big downturn in employment, could be one of the few broad-based indicators of the health of the labor market in September if the government shutdown drags on. U.S. statistical agencies, including the Bureau of Labor Statistics, won’t be producing economic releases for the duration of the shutdown, which began at 12:01 a.m. on Wednesday.

That means it is unlikely that the jobs report will be published on Friday. The weekly release of data on claims for unemployment benefits has already been suspended. If the shutdown continues, Federal Reserve officials may not have updated government employment data ahead of their Oct. 28-29 policy meeting.

The surprising contraction in monthly employment when growth was expected was due, in part, to the fact that ADP conducted an annual preliminary benchmarking of its payroll data. It was based on full-year 2024 figures on the distribution of employment across industries in the Quarterly Census of Employment and Wages, data produced by the BLS using unemployment-claims records.

“This recalibration resulted in a reduction of 43,000 jobs in September compared to pre-benchmarked data. The trend was unchanged; job creation continued to lose momentum across most sectors,” the company said.

September is a “special month” in terms of ADP’s statistical methodology, Nela Richardson, ADP’s chief economist, said Wednesday. She categorized the monthly private payrolls number as a ”projection” of what the QCEW data, which has a six-month lag, will say. The benchmarking process, which is similar to that undertaken by the BLS, helps to right-size the data.

“We found that once we benchmark that the data actually shows a September slowdown that has been consistent with what we’ve been reporting all year,” Richardson said. “In fact, though the numbers changed, the story and the narrative and the trend remain the same: Hiring momentum has slowed from the beginning of the year through September.”

While ADP published its revised August and September payroll data, the numbers are only based on a preliminary re-benchmarking. ADP has previously reported negative monthly job growth this year, but those numbers haven’t been quite this bad, and are only based on unrevised data. Revised numbers for all months won’t be ready until February.

The big picture is that Wednesday’s gloomy labor data is “far from definitive,” writes Oliver Allen, senior U.S. economist at Pantheon Macroeconomics.

“The September ADP report probably will receive more attention than usual, given that the government shutdown looks set to delay the publication of the official employment report,” he said. “But we still think that the ADP provides very little information of value.”

He said that by his estimate, ADP has been a far from perfect predictor of how the monthly government employment data would turn out. ADP has issued revisions as large as 348,000 since the report’s methodology was overhauled in August 2022, Allen wrote.

Additionally, other employment indicators aren’t yet pointing to a major labor contraction. The latest figures on jobless claims, issued for the week ended Sept. 20, didn’t show a significant jump. Researchers at the Federal Reserve Bank of Chicago also calculate that the unemployment rate remained steady as of mid-September and WARN notices, or disclosures of planned job cuts, have been range-bound.

The Institute of Supply Management’s Manufacturing Purchasing Managers’ Index report, published Wednesday, showed that employment ticked up slightly in the sector for September. The report’s employment index registered 45.3 in September, 1.5 percentage points higher than August’s reading of 43.8.

Still, the payroll drop was enough to make investors more confident that the Fed will cut interest rates later this month. Prices of interest-rate futures imply nearly 100% odds of a quarter percentage-point cut, according to the CME FedWatch Tool, compared with 96% on Tuesday.

The loss of jobs was fairly broad-based. Small and midsize companies reported that their workforces shrank, while employers with more than 500 staff reported a gain of 33,000 jobs.

The education and health services sector, once again, led the pack with job creation, adding 33,000 positions last month. The information sector added 3,000, while natural resources and mining added 4,000.

But employment in every other sector declined. Companies producing goods lost about 3,000 jobs and the services sector shed 28,000.

Healthcare has been one of the “stalwarts” of the labor market, Richardson said, a trend visible in both the ADP and BLS payroll data throughout the year. But Richardson says that is due more to demographics than labor conditions, given that the need for healthcare has grown as the population ages.

Pay gains sped up slightly with a gain of 4.5% year over year in September for those who stayed with their current employer, compared with 4.4% wage growth in August. Pay gains for workers who changed jobs slowed to 6.6% year over year in September, from 7.1% in August.

“What these numbers tell me for September is that wages remain robust. There’s still some tightness in the labor market,” Richardson said. “This is still a very stable labor market, but the hiring momentum has slowed.”

Write to Megan Leonhardt at megan.leonhardt@barrons.com